Workshops

Digital Health Technologies and AI for Studying and Predicting Episodic Health Events

Organizer: Ryan McGinnis

Presenters: Brett Meyer (Medidata Systems, brett.meyer@3ds.com ), Thurmon Lockhart (School of Biological and Health Systems Engineering, Arizona State University, Thurmon.Lockhart@asu.edu), Ryan McGinnis (School of Medicine, Wake Forest University, ryan.mcginnis@advocatehealth.org), Oguz Akbilgic (School of Medicine, Wake Forest University, oguz.akbilgic@advocatehealth.org), Asim Gazi (School of Engineering and Applied Sciences, Harvard University, agazi@seas.harvard.edu), Jiaee Cheong (Depatrment of Psychiatry, Harvard University, jcheong1@bidmc.harvard.edu), (representing PI John Torous), Megan O’Brien (Feinberg School of Medicine, Northwestern University, mobrien@ricres.org) (representing PI Arun Jayaraman), Sameer Neupane (Department of Computer Science, The University of Memphis, Sameer.Neupane@memphis.edu) (representing PI Santosh Kumar), Sina Masoumi Shahrbabak (Department of Mechanical Engineering, University of Maryland, smasoumi@umd.edu) (representing PI Jin-Oh Hahn), Varun Mishra (Khoury College of Computer Sciences, Northeastern University, v.mishra@northeastern.edu), Laura Barnes (Department of Systems and Information Engineering, University of Virginia, lb3dp@virginia.edu), Orson Xu (Department of Biomedical Informatics, Columbia University, xx2489@cumc.columbia.edu)

Digital health technologies (e.g., wearables) and associated study designs (e.g., ecological momentary assessments) provide a unique opportunity to study episodic health events such as panic attacks, emotional outbursts, depressive episodes, atrial fibrillations, freezing of gait episodes, seizures, migraines, falls, sleep apneas, cough, and scratch. Episodic health events span clinical domains and are notoriously difficult to study because they traditionally rely on patient self-reports of symptoms. The advent of digital health technologies have enabled more detailed studies of these important conditions with some already advancing from research to commercial implementation with FDA clearance. This workshop will bring together top researchers and their trainees who study episodic conditions with these technologies to share their approaches to analyzing and modeling these high frequency longitudinal time series data and the associated advances they have made in our understanding of these conditions.

From Bioelectronics to Digital Biomarkers: AI-Driven Continuous Health Monitoring

Organizers: Yayun Du, Yuanwen Jiang

Presenters: Nilanjan Sarkar, nilanjan.sarkar@vanderbilt.edu, Vanderbilt University; Yayun Du, yayun.du@vanderbilt.edu, Vanderbilt University; Yuanwen Jiang, ywjiang@seas.upenn.edu, University of Pennsylvania; Anthony Banks, tbanks@northwestern.edu, Northwestern University and Company Neurolux; Marco Rolandi, mrolandi@ucsc.edu, UC Santa Cruz

This interdisciplinary workshop brings together pioneers in hardware innovation, algorithm design, and translational medicine to explore the full continuum of AI-driven health monitoring. Attendees will gain insight into cutting-edge research spanning commercial-grade sensor fusion and affective computing (Dr. Sarkar), advanced multimodal bioelectronic sensing with human and animal validation (Dr. Du), next-generation implantable systems (Dr. Banks), neuromodulation and e-skin integration (Dr. Jiang), and programmable, AI-embedded therapeutic devices (Dr. Rolandi). Aligned with BHI 2025’s core tracks, the workshop fosters cross-disciplinary dialogue at the intersection of engineering, data science, and clinical practice—accelerating the path from biosensors to actionable digital biomarkers.

Coupled AI and Finite Element Methods: Discovering Novel Biomarkers from Complex Medical Datasets

Organizer: Nenad Filipovic

Presenters: Nenad Filipović (Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia, Bioengineering Research and Development Center (BioIRC), Kragujevac, Serbia, fica@kg.ac.rs); Dejan Krsmanović (CardioMed Technology Consultants, Miami, FL, USA dan@cardiomedtech.com); Tijana Geroski (Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia, Bioengineering Research and Development Center (BioIRC), Kragujevac, Serbia, tijanas@kg.ac.rs), Dejana Popovic (Mayo Clinic, Rochester, Minnesota, USA, Popovic.Dejana@mayo.edu)

Traditional biomarkers often capture static snapshots of biological processes, whereas AI coupled with FEM can reveal dynamic biomechanical signatures, tissue remodeling patterns, and functional connectivity biomarkers. The integration of heterogeneous medical data sources—including multi-modal imaging (MRI, CT, ultrasound), omics data (genomics, proteomics, metabolomics), wearable sensor streams, and longitudinal electronic health records—through physics-informed AI models enables the extraction of novel composite biomarkers with enhanced predictive power. The workshop will focus on computational methodologies for processing multi-dimensional medical datasets to discover biomarkers that capture the underlying physics of biological processes, addressing key challenges in biomarker validation, regulatory approval pathways, and clinical implementation.

From Sensors to Decision Making: Delivering Precision Biomarker Insights with Resource-Constrained AI

Organizer:Sevgi Zubeyde Gurbuz, Edgar Lobaton, and Ravi Chilukuri

Presenters:Dr. Edgar Lobaton, NC State University, Dr. Omer Inan, Georgia Tech and Biozen, Dr. Brinnae Bent, Duke University, Dr. Chenhan Xu, NC State University, Dr. Sevgi Z. Gurbuz, NC State University, Bill Kutsche, Murata, Bharath Rajagopalan, ST

Precision health monitoring requires an integration of sensing and computation with biomechanics and bioengineering; however, oftentimes these domains are treated independently. A complete system integrating these domains faces critical challenges involving real-time computation, sensor signal processing, and multi-modal sensing for precision biomarker estimation. This workshop brings together experts along this entire processing chain and includes both academic and industry perspectives relating to performance, price and real-world deployment constraints. The workshop will feature talks from seven experts:

    • Dr. Edgar Lobaton, NC State University, Acoustic Wearable Monitoring with Embedded AI
    • Dr. Omer Inan, Georgia Tech and Biozen, Designing CardioTag
    • Dr. Brinnae Bent, Duke University, Designing Intelligence Under Constraints
    • Dr. Chenhan Xu, NC State University, Precision Sensing in Human-centric IoT
    • Dr. Sevgi Z. Gurbuz, NC State University, Real-time AI/ML with RF Sensors
    • Kohta Nakai, Murata, RF Modules for Medical Devices
    • Swapnil Saha, ST, AI-enabled Sensors for Healthcare

The workshop will also feature two panel discussions related to the inter-disciplinary aspects of the workshop theme, spanning 1) sensing for precision health, 2) real-time edge computing, and 3) AI for sensing.

Workshop on Agentic AI for Healthcare: Ethics, Accountability, Resilience, and Trust (Agentic-HEART)

Organizer: Shahram Rahimi, Sudip Mittal, Noorbakhsh Amiri Golilarz, Subash Neupane

Presenters:

Dr. Vibhuti Gupta – University of Texas Medical Branch
Dr. Subash Neupane – Meharry Medical College
Dr. Noorbakhsh Amiri Golilarz – The University of Alabama
Dr. Hassan Al Khatib – Mississippi State University

The future of healthcare hinges on bridging the gap between groundbreaking innovation and systemic inefficiencies. Clinicians face mounting pressure to interpret siloed, multimodal data (imaging, genomics, EHRs) while managing time-sensitive decisions, a challenge compounded by voluminous medical knowledge. Agentic AI offers a transformative leap: autonomous systems that enhance diagnostics through AI-driven image analysis, personalize treatments via genomic and clinical data synthesis, and optimize workflows by automating administrative tasks like scheduling and billing. These agents can guide robotic surgery, predict patient risks, and deliver tailored patient support, improving both care quality and operational efficiency. Yet, ethical governance, seamless EMR integration, and clinician trust remain critical barriers to adoption. Through case studies and collaborative dialogue, we aim to accelerate the transition from theoretical potential to real-world impact, empowering stakeholders to build Agentic-AI systems that prioritize patient safety, clinician autonomy, and scalable healthcare transformation.


PROTECT-CHILD: Federated Intelligence for Secure and Personalised Paediatric Transplant Care

Organizer: Maria F. Cabrera, May D. Wang

Presenters: María F. Cabrera, Eugenio Gaeta, Pablo Lapunzina, Matteo Gabetta, Alessio Fioravanti, May D. Wang

Children undergoing organ transplants face unique long-term health challenges due to treatment-related complications and variability in immunosuppressive responses. The PROTECT-CHILD initiative aims to transform paediatric transplant care through data-driven, personalised approaches by integrating genomics, real-world data, and advanced computational tools. The project builds on the European Health Data Space (EHDS) vision, ensuring privacy, trust, and interoperability across borders. Leveraging the TransplantChild ERN and cutting-edge technologies like federated learning, zero trust architectures, OMOP on FHIR, and quantum algorithms, PROTECT-CHILD is setting a new standard in ethical, secure data sharing. This workshop will present the multidisciplinary efforts shaping the project, discuss the technologies and challenges involved, and explore future pathways for federated, privacy-preserving, personalised paediatric care.

Biomedical STAR-AI Workshop

Organizers: May D. Wang, Kathy Grise

Kathy Grise

Presenters:Amir Amini, Jonathan Beus, Maria F. Cabrera, Vince D. Calhoun, Rosa Chan, Daniel L. Drane, Dimitrios Fotiadis, Hassan Ghasemzadeh, Wayne Liang, Denise Lo, Bobak Mortazavi, Naveen Muthu, Tayo Obafemi-Ajayi, Peng Qiu, Md Mobashir Hasan Shandhi, Puneet Sharma, Wenqi Shi, Saurabh Sinha, Zachary West

Over the last decade, foundational artificial intelligence (AI) algorithms accelerate drug discovery and genomics, identifying therapeutic targets and personalizing treatments at unprecedented speed. In clinical practice, applied AI systems analyze multimodal biomedical data, predict patient risks and outcomes in real time, and optimize treatment plans for complex diseases. Moreover, by automating administrative tasks like documentation and scheduling, AI has the potential to alleviate clinician burnout and allow providers to focus more on patient care. However, the promise of AI to revolutionize medicine remains largely unrealized because the fragmented AI lifecycle has not been addressing the interdependent challenges of safety, trust, clinical utility, and ethical oversight.  Thus, BME department of Georgia Institute of Technology and Emory University has started pursuit of research and education in Biomedical Safe, Trustworthy, Actionable, and Responsible AI.

In this workshop, we aim to discuss AI implementation science that leads to fast growing area of Biomedical STAR-AI.

Presenter: Dr. Amir Amini

Title: Developing Deep Learning Models for Medical Imaging: Managing the Paucity of Training Data

STAR-AI Theme: Multimodal AI

Bio:

Amir A. Amini is Endowed Chair in Bioimaging and Professor of Electrical and Computer Engineering at the University of Louisville. His prior faculty appointments were at Yale and Washington University in St. Louis. He has had leadership roles in organization of numerous conferences in medical imaging and image analysis as scientific program committee member, scientific program chair, as well as conference chair, and was symposium co-chair of SPIE Medical Imaging in 2007 and the IEEE International Symposium in Biomedical Imaging in 2018. He has served as Associate Editor for IEEE Transactions on Medical Imaging, IEEE Trans. On Biomedical Engineering, IEEE Reviews in Biomedical Engineering, IEEE Open Journal of Engineering in Medicine and Biology, and Computerized Medical Imaging and Graphics.  He served as Vice President for Publications for the IEEE Engineering in Medicine and Biology Society in 2020-21. He was appointed Editor-in-Chief of IEEE Trans. on Biomedical Engineering as of 2025.

Under funding from the NIH, NSF, private foundations, and industry, his laboratory has conducted research in development and application of MRI methods for motion and flow measurement and is developing biomedical image analysis methods based on Deep Learning applied to a variety of imaging modalities with application to computer aided diagnosis, and radiation therapy of lung cancer.  He received the UMASS/Amherst College of Engineering Distinguished Alumni Award in 2020. He was elected a Fellow of the IEEE in 2007, to the College of Fellows of the American Institute for Medical and Biological Engineering in 2017, SPIE, the International Society of Optics and Photonics, in 2018, the Asia-Pacific Artificial Intelligence Association in 2021, and IAMBE, the international Academy of Medical and Biological Engineering in 2024.

Presenter: Dr. Jonathan Beus

Title: Getting AI to the Bedside

STAR-AI Theme: Multimodal AI

Bio:

  • Medical Director, Children’s Health Informatics Core Collaboration, Children’s Healthcare of Atlanta
  • Assistant Professor of Pediatrics, Pediatric Hospital Medicine
  • His primary operational and academic interests include secondary use of clinical data for research and quality improvement as well as the application of clinical decision support, working to bring AI to the bedside to prevent patient harm such as line infections and sepsis.
  • He also serves as the Children’s Healthcare of Atlanta informatics lead for PEDSnet, the 13-member national PCORnet® Clinical Research Network

Presenter: Dr. Maria Fernanda Cabrera

Title: From Innovation to Implementation: Lessons from Bringing AI into Real-World ALS and MS Care

STAR-AI Theme: Multimodal AI

Bio:

  • Professor of Biomedical Engineering in the Department of Photonic Technology and Bioengineering at the Universidad Politecnica de Madrid
  • Director for the Life Supporting Technologies Research Group (LifeSTech)
  • Secretary of the Active Ageing Association
  • Treasurer of the International Federation of Medical and Biological Engineering (IFMBE) and the International Uniion for Physical and Engineering Science of Medicine (IUPESM)
  • She is national expert for the European ACCESSIBLE EU resource centre on accessibility. Dr. Cabrera-Umpierrez works as a coordinator and PI on competitive research projects funded by both the European Union and Spanish institutions.
  • Her expertise spans a wide range of applications in the field of Information and Communication Technologies (ICT), applied to sectors such as health, social inclusion, and cultural heritage. This includes the personalization of services, human-machine interaction, interaction design, mobility and accessibility.
  • Dr. Cabrera-Umpierrez has authored more than 100 scientific articles published in national and international journals.

Presenter: Dr. Vince D. Calhoun

Title: Maximizing information in neuroimaging analysis: Flexible approaches for analysis and visualization

STAR-AI Theme: Trustworthy AI, Multimodal AI

Bio:

  • Founding Director, Tri-Institutional TReNDS Center.
  • Distinguished University Professor, Georgia Tech / Georgia State / Emory.
  • Pioneer of group ICA (Independent Component Analysis) on MRI, PET, EEG, and other modalities with multimodal fusion methods.
  • Recipient, OHBM 2024 Glass Brain Lifetime Achievement Award.
  • IEEE, AAAS, ISMRM, ACNP, IAMBE, OHBM, AAIA, and AIMBE Fellow
  • 1200+ publications; >120,000 citations.

Dr. Calhoun advances Trustworthy AI by developing interpretable multimodal neuroimaging methods. His pioneering work in independent component analysis and brain connectivity has reshaped how neuroscience leverages AI, with an emphasis on reproducibility and fairness across diverse populations.

Presenter: Dr. Rosa Chan

STAR-AI Theme: Multimodal AI

Bio:

  • Professor in the Department of Electrical Engineering at City University of Hong Kong
  • Her research interests include computational neuroscience, neural prosthesis, and human-computer interaction.
  • She was the co-recipient of the Outstanding Paper Award of IEEE Transactions on Neural Systems and Rehabilitation Engineering in 2013, for research breakthroughs in mathematical modeling for cognitive prosthesis. She also shared the Gold Medal in Inventions Geneva Evaluation Days (IGED), a virtual edition of the International Exhibition of Inventions of Geneva, in 2021, for a youth sports education and management SaaS platform based on AIoT.

Presenter: Dr. Daniel Drane

Title:  Application of AI in a Clinical/Translational Research Laboratory in Neuroscience

STAR-AI Theme: Multimodal AI

Bio:

  • Professor, Departments of Neurology and Pediatrics, School of Medicine
  • Faculty, Children’s Center for Neurosciences Research (CCNR), Children’s Healthcare of Atlanta
  • Affiliate Professor of Neurology, School of Medicine, University of Washington
  • Faculty, Emory Epilepsy Center, Department of Neurology, School of Medicine

Presenter: Dr. Hassan Ghasemzadeh

Title: Beyond Prediction: Agentic Intervention Discovery

STAR-AI Theme: Multimodal AI

Bio:

  • Associate Professor, College of Health Solutions, ASU
  • Two decades of experience in digital health and AI
  • Leadership in interdisciplinary NIH- and NSF-funded projects on design, development, and deployment of health AI systems
  • Member of graduate faculty in biomedical informatics & data science, computer science, computer engineering, and biomedical engineering at ASU.
  • Received a prestigious National Science Foundation Early Career Research award on wearable-based health monitoring technology.
  • 10 years of industry experience
  • Secured >$14M in external funding

Presenter: Dr. Wayne Liang

STAR-AI Theme: Multimodal AI

Bio:

  • Director of Informatics Education & Outreach; Associate Professor
  • Associate Professor, Pediatrics, Emory University
  • Dr. Liang is particularly interested in clinical informatics and clinical research informatics for pediatrics, hematology, oncology and bone marrow transplant, as well as biomedical informatics education.
  • Board certified in Pediatrics, Pediatric Hematology/Oncology, and Clinical Informatics (ABPM), and is a Fellow of the American Medical Informatics Association (FAMIA).

Presenter: Dr. Denise Lo

STAR-AI Theme: Multimodal AI

Bio:

  • Associate Professor Emory University
  • Dr. Denise Lo joined the Emory Liver Transplant team in 2016. She completed her medical education at Northwestern University, General Surgery residency at Georgetown University, and fellowship in Abdominal Transplant Surgery at Emory. Dr. Lo specializes in liver transplantation and hepatobiliary surgery. She is also the Associate Chief Medical Officer at Emory University Hospital. Her primary academic interest is the science of high-quality healthcare delivery.

Presenter: Dr. Bobak Mortazavi

Title: Multimodal Representations and Real-Time Cardiovascular Monitoring

STAR-AI Theme: Multimodal AI

Bio:

  • Associate Professor in the Department of Computer Science & Engineering at Texas A&M University
  • His research focuses on systems and analytics for personalized health, emphasizing machine learning methods that enhance clinical outcomes through multimodal modeling and dynamic risk prediction.
  • His work has pioneered the integration of machine learning and embedded systems with clinical outcomes research, including clinical risk prediction models and models for individual monitoring through passive sensing in remote, free-living environments.

Presenter: Dr. Naveen Muthu

STAR-AI Theme: Multimodal AI

Bio:

  • Director of Health Informatics Core Innovation Services, Children’s Healthcare of Atlanta,
  • Assistant Professor of Pediatrics at Emory University, Pediatric Hospitalist
  • Naveen co-directs AI strategy in Children’s Healthcare of Atlanta. He has been involved in multiple predictive model implementations and evaluations.
  • He also co-directs the Pediatric Clinical Decision Support Collaborative, a consortium of 11 pediatric health systems aiming to leverage CDS to improve quality and safety in pediatric healthcare.

Presenter: Dr. Tayo Obafemi-Ajayi

Title: Ethics vs. Regulation: Converging Frameworks for Trustworthy Human-Centered AI in Biomedical Research

STAR-AI Theme: Multimodal AI

Bio:

Tayo Obafemi-Ajayi PhD,  is an associate professor of Electrical Engineering (Guy Mace Professor of Engineering) at Missouri State University (MSU) in the Engineering Program, a joint program with Missouri University of Science and Technology, Rolla, Missouri. She received the MSU Atwood Excellence in Research and Teaching award 2924 and Board of Governors Faculty Excellence award 2025. She serves as chair of IEEE CIS Technical Committee on Ethical, Legal, Social, Environmental and Human Dimensions of AI/CI (SHIELD). She is also a Technical Representative on the Administrative committee of IEEE Engineering Medicine and Biology Society.  Her research focus on developing explainable and ethical machine learning/AI algorithms for broad utility in biomedical applications.

Presenter: Dr. Peng Qiu

Title: Bulk RNA-seq Deconvolution via a U-Net Based Deep Learning Model

STAR-AI Theme: Multimodal AI

Bio:

  • Wallace H. Coulter Distinguished Faculty Fellow
  • Peng Qiu is a professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University. His research interests are in the areas of bioinformatics, machine learning, data integration, statistical signal processing, with applications in genomics, single-cell analysis, and cancer.

Presenter: Dr. Mobashir Shandhi

Title: Developing Physiology-Informed, Reliable, and Clinically-Validated Digital Health Technologies

STAR-AI Theme: Multimodal AI

Bio:

  • Assistant Professor, School of Electrical, Computer and Energy Engineering, Biodesign Institute, and School of Medicine and Medical Engineering, ASU
  • His research interests are developing reliable and equitable digital health technologies (wearable sensors and machine learning algorithms) that can enable personalized health care and remote patient monitoring for patients with chronic and infectious diseases.
  • He has a particular focus on developing these digital health technologies that can provide a low-cost alternative to current diagnostic and monitoring devices used in the clinic and remote home monitoring, which can ultimately be used in limited-resource settings, improve access to health care, and reduce health-related disparities.
  • He is a recent recipient of the American Heart Association Career Development Award.

Presenter: Dr. Puneet Sharma

Title: Getting AI to the Bedside

STAR-AI Theme: Multimodal AI

Bio:

  • Assistant Professor of Pediatrics, Emory University School of Medicine, Division of Neonatology
  • His research, which has been supported by both the National Institutes of Health and American Academy of Pediatrics, focuses on the use of artificial intelligence to improve diagnostic and treatment strategies for critically ill newborns.
  • He is especially focused on the development of imaging and multi-modal AI for critically ill newborns, and this work was featured as an Editor’s Choice presentation at TEDx Atlanta 2025.

Presenter: Dr. Wenqi Shi

Title: Accelerating Clinical Research and Biomedical Discovery with Agentic Reasoning in LLMs

STAR-AI Theme: LLMs in Healthcare

Bio:

  • Assistant Professor in the Department of Health Data Science and Biostatistics at UT Southwestern
  • Dr. Shi’s research interest lies at the intersection of artificial intelligence (AI) and healthcare, advancing both fundamental algorithms and applied systems for precision and personalized medicine. With a dedicated focus on pediatric healthcare, cancer, and rare diseases, she actively works on developing large language models (LLMs) for translational medicine, advancing agentic AI for biomedical discovery, and promoting responsible AI practices to improve real-world clinical research and practice.

Presenter: Dr. Saurabh Sinha

Title: Spatial Transcriptomics and the Art of Differential Expression

STAR-AI Theme: Multimodal AI

Bio:

  • Wallace H. Coulter Distinguished Chair in Biomedical Engineering
  • Saurabh Sinha received his Ph.D. in Computer Science from the University of Washington, Seattle, in 2002, and after post-doctoral work at the Rockefeller University with Eric Siggia, he joined the faculty of the University of Illinois, Urbana-Champaign, in 2005, where he held the positions of Founder Professor in Computer Science and Director of Computational Genomics in the Carl R. Woese Institute for Genomic Biology until 2022.
  • He joined Georgia Institute of Technology in 2022, as Wallace H. Coulter Distinguished Chair in Biomedical Engineering, with joint appointments in Biomedical Engineering and Industrial & Systems Engineering. Sinha’s research is in the area of bioinformatics, with a focus on regulatory genomics and systems biology.
  • Sinha is an NSF CAREER award recipient and has been funded by NIH, NSF and USDA. He co-directed an NIH BD2K Center of Excellence and was a thrust lead in the NSF AI Institute at UIUC. He led the educational program of the Mayo Clinic-University of Illinois Alliance, and co-led data science education for the Carle Illinois College of Medicine.
  • Sinha has served as Program co-Chair of the annual RECOMB Regulatory and Systems Genomics conference and served on the Board of Directors for the International Society for Computational Biology (2018-2021). He was a recipient of the University Scholar award of the University of Illinois, and selected as a Fellow of the AIMBE in 2018.

Presenter: Dr. Zachary West

Title: Getting AI to the Bedside

STAR-AI Theme: Multimodal AI

Bio:

  • Fellow, Emory University School of Medicine Department of Pediatrics, Division of Cardiology
  • His work bridges clinical care and artificial intelligence for patients with cardiac conditions, focusing on integrating predictive analytics into real-time hospital workflows to improve patient outcomes
  • He leads the development and implementation of a machine learning algorithm designed to detect early clinical deterioration in pediatric cardiology patients, having demonstrated that the late recognition of these patients worsens mortality